I am a junior faculty member in the Department of Critical Care Medicine at the University of Pittsburgh. I am trained in Pulmonary and Critical Care Medicine and have a Masters of Health Science degree in clinical research. My short-term goal is to become expertly trained in clinical research of the critically ill, specifically focusing on the integration of emerging knowledge of the biology of critical illness with advances in clinical research, biostatistics, and data management. My long-term goal is to bring a new level to critical care medicine where an individual's pathophysiologic response to illness is used to predict outcome and guide care. Under the tutelage and mentorship of Dr. Angus (sponsor) and key consultants, I have developed a set of formal coursework, directed reading and tutorials that will form the educational and training core necessary to achieve my goals. This training and education will take place in the Department of Critical Care Medicine, the Department of Biostatistics in the Graduate School of Public Health, the Center for Human Genetics and integrative Biology at the University of Pittsburgh Medical Center, and at the adjacent Carnegie Mellon University. My research project, "Microarray Determinants in Community-Acquired Pneumonia (CAP)" is designed to both provide important new information and provide an important educational opportunity complimenting the other components of my training and education. CAP is a major public health problem, frequently fatal in the elderly. The proposed study will perform a high throughput genetic analysis of elderly Caucasian males with pneumococcal CAP using microarray technology. Two substudies will occur. The first substudy is designed to develop a mortality risk prediction that distinguishes survivors and non-survivors, who are phenotypically similar, from a pre-existing database of stored whole blood samples (n=50). The second substudy involves the recruitment of a prospective inception cohort (n=50) of elderly white males with pneumococcal pneumonia. The value that provocation testing (with pneumococcal vaccine) has in providing additional information with regard to predicting clinical outcome will be explored. This second substudy will also serve as a validation cohort for the model developed in the first substudy. This proposed study will generate a new predictive tool of clinical outcome based on gene expression analysis in the narrowly defined, homogeneous cohort of elderly Caucasian males with CAP and potentially identify genes not previously known to be involved in the immune response to infection. Information will also be generated regarding the pathophysiology of infection and possible points of early intervention with medical therapy that may alter the course of the disease process. [unreadable] [unreadable]